Facial Emotion Classifier (FCE) is a facial emotion recognition algorithm created using python to classify facial emotions. The algorithm can identify six emotions: anger, fear, disgust, surprise, neutral, and happiness. The algorithm was trained on Kaggle's Face Expression Recognition Dataset. First, the data was preprocessed by overlaying images with landmarks and then randomly rotating each image between [-30, 30] degrees. Then, the images were normalized using Pytorch's batch normalization. The model was trained for 50 epochs, with a learning rate of 1e-3, a weight decay of 1e-5, and a batch size of 128 using the Adam optimizer. After finishing training, the model performed with an accuracy of 0.535 on the test set.
Created using Python 3.8, Pytorch, and open-cv.
ChatUM (pronounced Chat'em) was a social media website created in coordination with others to allow for student communication during COVID-19. At the time, classes were remote so students did not have many avenues to socialize. This social media platform provided the ability to create accounts, posts, and comments that were customizable to each student and class. Additionally, cross-class socialization was supported by allowing students to create a main page for their feed so they can see posts from several classes at once.
Created using React, Firebase, Javascript, HTML, and CSS.
ChefU was a crossplatform (IOS/Android) mobile app that allowed users to create and share recipes. It included features such as supporting video snippets for each separate step in a recipe as well as custom built-in timers that were preset in accordance with whatever a recipe demanded. Furthermore, with accessability in mind, the app also has support for text-to-speech and speech-to-text for the impaired.
Created using Flutter, Firebase, and Dart
